Post reply

Warning: this topic has not been posted in for at least 120 days.
Unless you're sure you want to reply, please consider starting a new topic.

Note: this post will not display until it's been approved by a moderator.

Name:
Email:
Subject:
Message icon:
Verification:
Sum of two plus two?:

shortcuts: hit alt+s to submit/post or alt+p to preview


Topic Summary

Posted by: Admin
« on: January 25, 2013, 16:22:10 »

1) 64 bit version will be released very shortly as is new Mac version, 1-2 weeks.
2) Main problem with Linux is GUI, all other parts are universal.
3) Player will be added in future versions.
4) You don't need to open them, all groups already shown grouped. See same gray color and number on the right. You need open them only to see which files exactly duplicates for what, this situation appears rarely in huge groups or error groups, in most cases you already see grouping. More detailed:
It's our mistake, we just don't released new articles for 1.8.x grouping in this version grouping different from 1.7.x, old articles only misguide you (we didn't changed them because wait for Mac version to write article for all version at once). Example:
we have duplicate pairs:
1.mp3 - 2.mp3 - 95%
1.mp3 - 3.mp3 - 70%
2.mp3 - 3.mp3 - 80%
You see in result tab 1 group:
1 95% = max(95, 70)
2 95% = max(95, 80)
3 80% = max(70, 80)
It's all you need you have full information, you only need collapse elements to see exactly values for percentage and if group doesn't full.
4) About DirectShow this decoder can decode anything just need 3rd party decoder, WAV files opened via old ACM decoder (via DirectShow too)
5) Yes we improve image comparing algorithm in future
Posted by: dksvertix
« on: January 24, 2013, 22:45:22 »

* 64-bit was one suggestion, since, when processing 170 000+ files, Similarity may run out of RAM on 32-bit and crash, loosing all work progress. But you are already at it, that's good.
Or maybe you can implement advanced memory management into 32-bit too, so processing 170 000+ files would be also possible on 32-bit OSs? I haven't tried yet, since I do think that it will run out of RAM and crash.

(* Linux (Ubuntu and Debian) support wouldn't be bad either if you are already building for Mac.)

* Built-in player
Yes, we need that! Current method of opening end-user's own player, even on a powerful enough PC, the player will open with a delay and sometimes with longer delay and in worst cases, it won't respond and won't open at all, making me mark wrong file and I always get to listen the beginning of the song and have to manually seek to the middle. And that all wastes our time. A built-in player will fix that.
Built-in player should be just like in Audio Comparer, but better. Play, pause, start from the beginning. Continue playing the next song from the same %, yes percentage, of previously played song. Allow to click on the trackslider to quickly change play from position and don't force or even allow to drag it, since that makes player play horrible noise.

* Remove that horrible collapsed tree view from under both results tabs!
In Audio Comparer, there is at least a function which automatically expands them all, so I wouldn't have to spend half of my life on expanding them all manually.
But what I really would like, is to get rid of this horrific, agonizing collapsed tree view, since you have no idea, how much of an eyesore it is and, how much time it wastes!
Take a look, how Duplicate Cleaner displays its results, it's beautiful! No eyesore, no idiotic collapsed trees! But that's because Duplicate Cleaner's results displaying algorithm is a bit more smarter / user friendlier too, which I will try to cover in the next suggestion.
So, no collapsed trees, each duplicate group is separated from another group with just a bit different, but highly noticeable background color.

* Get rid of duplicate duplicate groups
I read your article "Working with groups in Similarity" and that was still not convincing enough for me to waste my time on up to 4 duplicate duplicate groups. If 1=2=3, then, why must you make separate groups with 2=1=3, 3=1=2, 1=2 etc (yes, I fed it with 4000 songs, which had a lot of dupes and this kind idiocy really was, what I saw in the results list), why, if the program already knows that 1=2=3 anyway? Do it like they do in Duplicate Cleaner: duplicates and similar to each other are in one group only! 1=2=3 and that's it! There is no highlighted "main" file to compare against, there is no reason for that idiotic collapsed treeview.

* Chiptune and other audio comparison support
MID, XM, MOD, IT, S3M, 669, MTM and STM and RAW.
I see it already seems to support WAV. A sentence: "Any audio-formats supported by Windows (through the DirectShow and Microsoft Media Foundation technology)" isn't really that informative.

(* Make duplicate image finder to work just as VisiPics, but better.)


I thought there was "keep me logged in forever" set, but somewhy, my login still got expired while I as writing this post, hmm...